One hundred and twenty-seven male patients were subjected to antiulcer surgery for duodenal ulcer resistant to H2-receptor antagonist treatment. Fifty-four (group A) had been on conservative treatment for up to 6 months, while the remaining 73 (group B) had been on conservative treatment for more than 6 and up to 20 months. Of the group A, 43 underwent truncal vagotomy with pyloroplasty (group A1) and 11 highly selective vagotomy (group A2). Of group B, 52 underwent truncal vagotomy with pyloroplasty (group B1) and 21 highly selective vagotomy (group B2). Follow-up ranged between 18 and 72 months (mean 37 months). There were one ulcer recurrence in group A1, none in group A2, nine in group B1 and five in group B2, the difference between group A and group B being statistically significant (p less than 0.05). There was significantly higher nonulcer-associated morbidity after truncal than after highly selective vagotomy (p less than 0.05). No significant difference in the degree of peak acid output reduction was observed between the patients with and those without ulcer recurrence. These findings show that the administration of H2-receptor antagonists for more than 6 months in duodenal ulcer patients who, however, fail to have their ulcer healed is associated with high recurrence rate after vagotomy. It is suggested that such patients should undergo vagotomy as soon as they fulfill the criteria of resistance to H2-receptor antagonists. If conservative treatment has lasted for more than 6 months, vagotomy plus antrectomy has to be considered as the surgical treatment for these patients, with the possible cost of higher nonulcer-associated morbidity.